Ellydee

This app is not a frontier model, and anyone who judges this app against ChatGPT or Claude is making a fundamentally flawed comparison. Ellydee hosts a few popular models and adds its own flare via fine-tunes. You get a little more control over the model than the frontier models as well, because they allow you to change the advanced parameters of the model you’re using which is really important since ai companies are increasingly trying to control what its users are allowed to do. The app isn’t perfect, but from my understanding, Ellydee.ai is a pretty small company and the devs seem to run it as a passion project. It definitely has its glitches at times, and it doesn’t feel quite as streamlined as ChatGPT or Claude or the like, but I’ve gotten used to it because I like Ellydee’s vision. The devs are active on discord, so at least as of right now, it’s a smaller community and if you give them feedback in discord, they respond pretty quickly which is a really cool experience. I don’t know a single AI company that interacts with its users in this way. The mission of offering an unrestricted LLM (everything is generally allowed up until you get into the really dark/illegal/evil category) is very noble and really important. I personally use Ellydee for anxiety support and it’s been so great. It sucks that we live in a world where sometimes, when you’re talking to a frontier model and you’re really anxious or depressed, they model will just flat out refuse to engage, or start sending you BS recommendations to call a mental health line. It’s honestly insulting to users to do that, but the Brightside model in ellydee doesn’t blink and will help me work through my anxiety without sending refusals. The commitments to privacy of user data and ecologically sustainability aren’t the reason I subscribed but are definitely a plus. Of course, my biggest fear would be that a corporate AI lab would try to buy them out, which would no doubt lead to them shutting down their anti-censorship model,. The other fear is that the devs start to drift from true freedom of speech (not including the most evil parts of humanity, I.e. intent to harm, etc. which should always be shut down)—and what I mean is going from allowing the user to write a story about a murder mystery, and then a year from now drifting to “well we can talk about murder mystery, but you can’t talk about specific methods of the murder” and then its a slippery slope into the same censorship every other company has. But Ive not seen any evidence of this happening, just discussing what the risks of this company might be. Overall I really love the concept, and I have become very fond of the models on here. Their entire vision is something that all reasonable people align with, the all just needs a little polish because it can be rough or glitchy at times. But that comes with time as their business grows. I wish the devs the best in staying committed to their principles and their user base!

I think this chatbot is more sympathetic and empathetic than ChatGPT and has less restrictions but it revolves around tokens. I paid for a $10 monthly subscription. After using it for about a week, I got this error. “AL_APICallError: Insufficient Balance” Turns out my tokens have ran out and I may not be able to use the AI again until next month. I could pay more money for a higher subscription to get more tokens but they run out so quickly. And there’s no option where you get unlimited tokens to use. You don’t even get a token counter on the app to see how many credits you’re using. It just cuts you off when you’ve reached your limit. That’s what really downgrades the entire app for me. If it wasn’t for that, I would highly recommend this. But as is, I’m not sure if this is worth your money